The use of digital breast tomosynthesis (DBT) as surveillance to detect recurrence among women with a history of breast cancer was found to decrease the rate of false positives with high specificity, but did not lead to an improvement in detection when compared with digital mammography, according to the results of a study published in Radiology.1
“To our knowledge, this study is the largest to date of breast cancer survivors who underwent post treatment surveillance with DBT,” the study authors wrote.
Currently, guidelines recommend surveillance for women who underwent treatment for breast cancer for the early identification of locoregional recurrence or new ipsilateral or contralateral breast cancers. In an accompanying editorial, Regina Hooley, MD, and Reni Butler, MD, of Yale School of Medicine, wrote, “No clear consensus exists about the optimal screening regimen for women with a previous breast cancer diagnosis.”2
The retrospective study included 8170 women with a history of breast cancer who underwent 9019 digital mammography or 22,887 DBT evaluations between 2008 and 2011 or 2013 and 2017, respectively.1 Breast imaging radiologists reviewed the mammograms with computer-aided detection.
The mean age was 64 years and the majority of women were White. Three percent were were Black; race was not specified for 7% of the cohort. The breast density of most women consisted of scattered areas of fibroglandular density (52%, digital mammography group vs 55%, DBT group) followed by ‘heterogenously dense” classification (39% vs 35%), almost entirely fatty (5% vs 7%), and extremely dense (4% vs 3%). Nearly all women had undergone a previous screening mammogram.
The abnormal interpretation rate, or false-positive rate, was significantly lower with DBT at 5.8% compared with 6.2% in the mammography group (odds ratio [OR], 0.80; 95% CI, 0.71-0.91; P =.001). DBT was also associated with greater specificity, with a rate of 95.0% vs 94.7% with mammography (OR, 1.23; 95% CI, 1.07-1.41; P =.003).
However, the improvements in interpretation and specificity seen with DBT did not translate to an improved detection rate. DBT resulted in a detection rate of 8.3 per 1000 examinations compared with 10.6 per 1000 examinations for mammography (OR, 0.76; 95% CI, 0.57-1.02; P =.07). The detection rate for invasive and in situ cancers was also similar between the 2 groups.
A total of 86 interval cancers were identified during surveillance. MRI screening detected 33% of interval cancers. Symptoms were present among 58%. Nine percent of interval cancers were detected by computed tomography and/or positron emission tomography, after mastopexy or mammoplasty, or after mastectomy.
“Given that the cancer detection rate did not significantly change after DBT integration and that one-third of interval cancers were detected at MRI, our results suggest that supplemental screening with MRI will continue to be useful in this high-risk population,” the study authors concluded.
Disclosures: Some of the study authors disclosed financial relationships with the pharmaceutical industry and/or the medical device industry. For a full list of disclosures, please refer to the original study.
- Bahl M, Mercaldo S, McCarthy AM, Lehman CD. Imaging surveillance of breast cancer survivors with digital mammography versus digital breast tomosynthesis. Radiology. Published online December 22, 2020. doi:10.1148/radiol.2020201854
- Hooley R, Butler R. Digital breast tomosynthesis may not provide optimal surveillance of breast cancer survivors. Radiology. Published online December 22, 2020. doi:10.1148/radiol.2020204219